Essay: Aboveground-belowground biodiversity interactions/ possibility of managing soil life underneath crops

To date, interactions between crop and soil biodiversity are still not well predictable, but the idea that changing aboveground diversity will change soil biota is increasingly recognized. Specific soil organisms called nematodes have an important function in soil food webs. This research focuses on the effect of diverse grass-clover mixtures on the occurrence of nematodes in the soil. Total nematodes and two functional groups of nematodes are taken into account. The two groups are entomopathogenic nematodes (EPN’s) of genera Heterorhabditis and Steinernema, and fungal feeding nematodes of the genera Diphtherophora, Aphelenchoididae and Filenchus. Nematodes in general can be used as bioindicator for soil ecological health; EPN’s parasitize on insects larvae and can have a function as biocontrol agents in agricultural fields; fungivorous nematodes are even more frequent than EPN’s and represent abundance of fungi. By taking soil samples and carrying out Q-PCR’s, nematodes in the soil were determined. All studied grass-clover plots are situated in the same field, yet have different treatments considering diversity in species composition and time since tilling. A higher total nematode abundance was found in the older grass-clover plots. Higher aboveground species diversity in grass-clover plots did not result in higher nematode abundances. The genera Steinernema, Diphtherophora and Filenchus were found to occur patchy. This study tries to contribute to the understanding of aboveground-belowground biodiversity interactions and the possibility of managing the soil life underneath our crops.
Introduction
Soil biodiversity is known to play a key role in crop productivity by providing ecosystem services as decomposition and nutrient cycling [1,2,3]. Current studies try to reveal and explain feedbacks between aboveground and belowground diversity. To date, the effect of interactions between crop and soil biodiversity is not well understood. However, the idea that changing aboveground diversity changes soil biota is increasingly recognized [4].
In the soil food web, nematodes play a crucial role [5]. They occur ubiquitous in all habitats. In agriculture nematodes are mostly known as plant parasites, while nematodes encompass a much larger variety which feed on e.g. bacteria, arthropods, other nematodes or fungi. For the reason that nematodes are divided in a broad range of trophic groups, they could be used as bioindicator of soil ecological health. The presence of nematodes from a specific trophic group represents the presence of their feed. [6]
The first group to be studied are the “entomopathogenic nematodes” (EPN’s). These nematodes of the genera Steinernema and Heterorhabditis parasitize on insect larvae [7]. EPN’s parasitizing harmful insects can be used as biocontrol agents [8,9]. For this reason, more knowledge on the occurrence of EPN’s could be useful for future agricultural systems. EPN’s occur in a broad range of terrestrial habitats [7].
Several studies throughout the world found divergent results considering the habitat and soil preference of Heterorhabditis and Steinernema. Mráček et al. [12] found EPN’s more abundant in forest habitats and light soils (sand/loam) rather than open vegetation and heavy soils (clay). Valadas et al. [13], isolated EPN’s both from forest soils and agricultural land. The preference of EPN’s for sandy/loamy soils is scientifically accepted [14]. Still, their presence in a certain habitat is not well understood.
The second focus group are the fungivorous nematodes. These fungal feeding nematodes are often more abundant in ecosystems later in the succession than in just established ones. Older ecosystems normally contain more soil organic matter and fungi which feed on this [15]. Therefore, organic matter is expected to correlate with fungal feeding nematodes.
The presence of nematodes in general is not well predictable, due to occurrence of nematodes being influenced by multiple abiotic and biotic factors such as available water, organic matter (OM) content [15] or host distribution [14]. Howard [5] describes that fungal and bacterial feeding nematodes increased after organic matter and irrigation was applied.
The hypothesis that nematodes are more abundant in perennial habitats is taken into consideration in this research. The effect of difference in time since tilling of the grass-clover mixtures (sown this year or last year) on nematode abundance will be examined.
Next to time since tilling, also the effect of above ground crop diversity, on total, fungivorous and entomopathogenic nematode abundance will be examined. Plant species differ in the effects they have on soil biota and the effects range from negligible to large [4]. Next to that, Altieri et al. [16] indicates that a more diverse cropping system has the ability to maintain stability, soil fertility, pest suppression and crop production.
This research
This study focuses on whether nematode communities are more abundant in diverse and older grass-clover mixtures than in less diverse and younger mixtures.
This way, the study tries to contribute to the understanding of aboveground-belowground biodiversity interactions and the possibility of managing the soil life underneath our crops.
Research questions
The objective of this study was to gain inside in the following questions:
1. What is the effect of the aboveground diversity of grass-clover mixtures on the presence of fungivorous, entomopathogenic and total amount of nematodes in the soil?
2. What is the effect of time since tilling of a grass-clover mixture on the entomopathogenic, fungivorous and total amount of nematodes in the soil?
The following hypotheses were formulated:
1. If more diversity aboveground leads to more abundance belowground, a more diverse grass-clover mixture will lead to a more abundant nematode community.
2. If more time since tilling for the crop leads to a more diverse and abundant soil life, a grass-clover mixture of 2 years old will lead to a more abundant nematode community than the mixture of 1 year old.
To test these hypotheses, soil samples were taken from the experimental field at Droevendaal Farm in grass-clover soil consisting of four treatments. The treatments take into account two variables: time since tilling the crop and species composition of the crop. The data on nematode abundances were analysed on differences between treatments, additional data on soil organic matter content (obtained from Dirk van Apeldoorn of former research on the experimental field) could help to understand differences in nematode occurrence which do not originate from diversity in grass-clover.
Table 1 Treatments Mix and No-mix with species composition
 
Materials and Methods
Site description
This study was conducted in The Netherlands on the Droevendaal Organic Experimental Farm of Wageningen University and Research Centre (51°59’28″N, 5°39’42″E). Soil texture is sandy loam. The research is part of the system experiment which is carried out by Dirk van Apeldoorn (started in 2014).
Apeldoorn’s research is about the effect of diversity in time, space and genotypes on yield and resilience of the farming system. A schematic overview of the experimental field is given in figure 1. The experimental field consists of 8 strips where seven different crops are grown, of which two strips of grass-clover. The strips are each divided in 12 plots.
Treatments
The first variable taken into account is species composition. Every plot in a grass-clover strip was sown with a seed-mixture either of treatment No-mix or treatment Mix. Species composition of treatments Mix and No-mix are indicated in table 1. Whether plots got treatment Mix or No-mix is indicated in figure 1.
Treatment No-Mix Mix
Species
 
 
 
 
 
 
 
 
 
 
 
 
 
* Only in strip 6 with youngest grass-clover mix
Red clover (Trifolium pratense variety: Lucrum) Red clover (Trifolium pratense variety: Lucrum)
Italian Ryegrass (Lolium multiflorum) variety: Sultano (tetraploid) Italian Ryegrass (Lolium multiflorum) variety: Sultano (tetraploid)
Perennial ryegrass (Lolium perenne) varieties Maurizio and Tomaso
White clover (Trifolium repens) varieties: Alice, Riesling and Jura
Berseem clover (Trifloium alexandrinum)*
Crimson clover (Trifolium incarnatum)*
Chicory (Cichorium intybus)*
Plantain (Plantago lanceolata)*
 
The other variable is time since tilling which is expressed in two treatments: 1 and 2 years since tilling. Before the crop was sown, the strip was tilled. Strip 1 was sown in April 2014. Strip 6 was first sown in October 2014 and over seeded in March 2015.
Pilot study
Before the actual research, a small pilot study was carried out to explore the presence of EPN’s in strip 1 and 6 in the field. Primers for Heterorhabditis and Steinernema were used. In each of the two grass-clover strips, six soil samples were taken and put together as one composite soil sample. Further methods were similar to the actual experiment.
Soil sampling protocol
24 composite soil samples were taken in strip 1 and 6 (see figure 1) on October 12 and 13 2015. Per grass-clover plot, 6 stratified random samples were taken, after removing the litter layer, at 0-20 cm deep (see figure 2). Soil samples were taken with a soil gouge (diameter 13 mm). All 6 samples of one plot formed a composite soil sample, which were mixed in one bag and used for nematode extraction.
Nematode extraction
Following the methods from the practical guide protocol of the course ‘nematodes as biocontrol agents’ [10], nematodes were extracted from the soil with the Oostenbrink elutriator. Nematodes were obtained in a jar with 100 mL water suspension. The nematode suspensions were stored at 4°C for 14 days. After this, samples were split: 50 mL was put in a coned glass tube and 50 mL was left in the glass jars for total nematode count with the microscope [10].
Nematodes in 5 mL suspension were counted two times under the microscope. The average per sample was calculated.
Lab analysis
Soil samples in the coned grass tubes were washed, concentrated and a lysis was carried out [10]. In this process a known amount of control DNA was included. Next, samples were diluted 100x and a quantitative PCR (Q-PCR) was performed [10]. Primers that were used are indicated in table 2.
With SYBR green, double-stranded DNA could be observed over time during the Q-PCR. An amplification curve was made and the cycle threshold (Ct) values were read from the curves. The Ct values give a measure for the amount of DNA from the specific nematode group which was present at start of the Q-PCR reaction. In this way data on number of nematodes could be obtained.
 
Table 2 Primers used for quantitative PCR with their nematode group
Primer Trophic group
Heterorhabditis Entomopathogenic
Steinernema
Aphelenchoididae Fungivorous
Diphtherophora
Filenchus group III
Universal All nematodes
 
Data analysis
The number of nematodes in a Q-PCR sample was calculated using a primer-specific calibration curve. The number of nematodes in the sample was corrected by the amount of control DNA that was lost in the process. The percentage of lost control DNA was assumed to be representative for the percentage of lost nematode DNA.
The numbers of fungivorous, entomopathogenic and total nematodes for grass-clover strip 1 and 6 were taken as separate datasets.
For every data-set, the difference between the nematode abundances in strip 1 and strip 6 was analysed. A Friedman test was used for non-normally distributed data and a two-way ANOVA for normally distributed data.
Possible correlations between nematode occurrence and organic matter percentages were explored using a Spearman’s rank order correlation coefficient. All statistical analyses were performed using R statistical software [11].
 
Results
Pilot study
The presence of the EPN genera Heterorhabditis and Steinernema was explored with a Q-PCR. Steinernema was shown to be present in the soil of the 2 year old grass-clover mixture, Heterorhabditis was not.
Entomopathogenic nematodes
Results of the Q-PCR on samples with Heterorhabditis and Steinernema primers were all zero. The results can be assumed to be representative, for a control experiment on the functioning of the used primers showed positive results.
Total amount of nematodes
Total nematode count data were not normally distributed, therefore a Friedman test was used to test for significance of differences between Mix and No-mix and between strip 1 and strip 6. The null hypothesis that nematodes are equally abundant in the 1 year old as the 2 years old grass-clover mixture could be rejected. P-value = 0.007, suggesting a difference in nematode abundance with more time since tilling (figure 3B). Moreover, values on nematode abundance in the one year old mixture, obtained from the Q-PCR, were also significantly different from the 2 year old grass-clover mixture (P-value = 0.005, see figure 2B). The data on total nematodes from the Q-PCR method were normally distributed, so a two-way ANOVA was used. Obtained total nematode abundances from Q-PCR and counting differed with approximately a factor 10 (see figure 3A and B).
Fungivorous nematodes
Of the 24 samples studied, only 3 samples were positive for Diphtherophora and only 1 for Filenchus group III. Filenchus group III was found in strip 1, plot 3. Diphtherophora was found two times in plot 1, both in strip 1 and 6. Diptherophora was also found in strip 1, plot 9. Aphelenchoididae was found in 14 samples. See appendix 3 for results on number of fungivorous nematodes. Data on number of Diphtherophora and Aphelenchoididae were not normally distributed. Therefore, the non-parametric Friedman test was used for these data-sets. No significant differences between treatments Mix and No-mix or between 1 and 2 year old grass-clover mixtures were found (P-values > 0.05).
Organic matter
No correlations were found between organic matter content and nematode numbers. For none of the tested nematode groups the Spearman’s test showed significant correlations (P-values > 0.05). See appendices 1 and 2 for the graphs on the relation between nematode numbers and organic matter content.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Discussion
Entomopathogenic nematodes
The entomopathogenic nematodes from the genera Steinernema are thought to be very patchy distributed in the experimental field. If this is true, this could explain that after a successful pilot study, no EPN´s are found in the analysis. This hypothesis is contradictory with the results Mráček et al. [12] found, were EPN occurrence was more ubiquitous than patchy. However, the study of Vladimír Půža et al. [17] found that patchiness of EPN’s differed per species. In future nematode analysis in the same field, patchiness and distribution of Steinernema could be determined. Therefore, more samples have to be taken, preferably also in other strips, apart from strip 1 and 6. The proposed follow-up study is beyond the scope of this research.
Heterorhabditis was not found in the pilot study, nor in the final analysis, therefore this genera is assumed not to occur in the studied field.
Patchiness
Similar to Steinernema, Filenchus group III and Diphtherophora were assumed to occur very patchy. This assumption is made, for their presence was detected, but only in a few plots in the field.
Species composition
No difference was found between Mix and No-mix treatments within strips. Although the plots differed in species composition, nematode abundances were not significantly influenced by this. This was possibly due to the fact that the experiment only started in 2014 and differences in soil condition, soil biota, fertility, etc. (if there were any) didn’t have the time to evolve yet. Expected is an increase in differences between Mix and No-mix plots over time. This is because those treatments (Mix as more diverse and No-mix as less diverse) will be implemented on every crop sown in the future on these plots.
Strip 6 had a more diverse Mix treatment than Strip 1 (as indicated in table 1). Effects of this more diverse grass-clover mixture are assumed to be negligible, for the difference between Mix and No-mix was insignificant and the sown species mixture of strip 6 had less time to establish than strip 1.
Total nematodes
The data on total nematode abundance obtained from counting and performing a Q-PCR differed with approximately a factor 10, the question rises whether both data-sets are representing reality. In this study, the data obtained from counting nematodes under the microscope do most likely represent reality better than data obtained from the Q-PCR. This is because less errors could be made in the counting method. The samples were well shaken while 5 mL suspension was taken. Possible explanations for the large difference in nematode numbers between the two quantification methods, are pipetting errors during dilution of samples, or a possible lack of shaking of PCR samples during pipetting. In both scenario’s, less nematodes are possibly transferred into the final Q-PCR tubes than assumed. Although the data cannot represent reality quantitatively, the data within a method can be used as relative values, for the samples are all treated in the same way.
As stated in the results, nematode abundances in the 2 year old grass-clover mixture were shown to be higher than in the one year old mixture. This was significant in both count data and Q-PCR data.
Organic matter content
The field is relatively heterogeneous considering organic matter (OM) content. Expected was to find a relation between OM and fungivorous nematode abundances [15]. However, no correlation exists between Aphelenchoididae or total nematode abundance and organic matter content, according to the results. Possible explanations could be the influence of factors such as Mix/No-mix treatment, water availability or compaction of the soil.
Conclusions
The most important finding is the increase of total nematode abundance in the two year old grass-clover crop. More time since tilling had the effect of increasing nematode numbers. The hypothesis that nematodes are more abundant in perennial crops, rather than annual crops is supported. However, more support is needed from other research on different nematode genera to confirm this hypothesis.
The diversity in species composition had no effect on nematode abundance, so other annual and perennial agricultural systems have to further examine the use of diverse crop mixtures for nematode communities.
Finally, the conclusion can be made that we indeed can manage soil life. This study shows, however, that links in (agro) ecological systems are not easy to predict and changes in crop diversity or soil conditions are not always linearly correlated with soil life. From former research, is clear that by changing aboveground diversity, e.g. by diversifying the cropping system, you could change belowground soil biota [4]. Mechanisms behind these aboveground-belowground effects should be studied to make use of the ability to manage soil life in cropping systems.